taming-transformers
taming-transformers copied to clipboard
CompVis
Taming Transformers for High-Resolution Image Synthesis
In line `365` and `366`, `codebook_dim` and `num_tokens` are not defined before assigned ,,, ` https://github.com/CompVis/taming-transformers/blob/24268930bf1dce879235a7fddd0b2355b84d7ea6/taming/modules/vqvae/quantize.py#L365 `
I've been trying to set up a Google Colab notebook to train on a custom dataset. Custom notebook here: https://colab.research.google.com/drive/1CqbxgNHUY56j96yC3flyCuSNCkR5XWPl (although you'd need your own custom data to test) I...
Thank you for the great work! I tried to repoduce `VQGAN OpenImages (f=8), 8192, GumbelQuantization` model based on the config file from the cloud. (the detailed config file is in...
Hello! Now I'm trying to generate some pictures by vqgan+clip on my laptop with 2GB VRAM. Obviously I get some errors of out of memory. That's about my question. What's...
Im trainning on a custom dataset and having some difficulty understanding how to properly track the trainning process. Could someone please elaborate a little bit on the differences between `loss`,...
[This commit ](https://github.com/CompVis/taming-transformers/commit/6347a7de08b169ba20c2c35595e9315e548c50c4) introduced new *.ckpt files, however the *.yaml config file isn't avaliable in the zip file and I couldn't find it anywhere else, will those config files be...
By default the images generated are based on their step count doubling each time for the first epoch, then for each subsequent epoch only the first step inputs and reconstructions...
I keep getting this error in WSL, even though I do have a GPU: ```Running on GPUs 0, Working with z of shape (1, 256, 16, 16) = 65536 dimensions....
Great paper! I am trying to retrain this model on an image dataset where I'm able to generate the segmentation masks using DeepLab v2. However, I don't have a config...
Hi, Thank you for your great work and repo. I need to use pretrained VQGAN models for FFHQ and CelebA-HQ datasets, separately. You already shared faceshq-vqgan, but i specifically need...
